Optimization of cut-offs for repeater chains

This repository stores the implementation of the algorithm introduced in Efficient optimization of cut-offs in quantum repeater chains by Boxi Li, Tim Coopmans and David Elkouss. It includes two implementations:

• The numerical algorithm calculating the waiting time distribution and the fidelity of the delivered entangled state.
• The optimizer used to optimize the cut-off time for maximal secret key rate.

Tutorial

A tutorial is written in Jupyter Notebook. By clicking the badge above, you will be directed to an online Jupyter Notebook. After the loading succeeds, please find and click tutorial.ipynb. Then you can run examples in the notebook online without installing anything! (The loading can take a few minutes. It takes too long, please try again half an hour later.)

Download

To download or clone the repository, using the green button Clone or download.

Prerequisites

The following Python packages are required for running the core algorithms:

NumPy, Scipy, Numba

In addition, we use Matplotlib for plotting and pytest for unit tests.

To install the packages, you can use

pip install numpy scipy numba matplotlib pytest

or

conda install numpy scipy numba matplotlib pytest

if you are using conda environment.

For GPU accelerated convolution, you will need

CuPy

See CuPy installation for details

File overview

• The protocol units such as entanglement swap, distillation or cut-off are defined in protocol_units.py and protocol_unit_efficient.py. The later makes use of customized convolution in Fourier space.
• The core code for the numerical simulation of repeater chains is under repeater_algorithm.py.
• The optimizer can be found in optimize_cutoff.py.
• Examples for computing repeater protocols and optimizing the cut-off time are given in examples.py
• The figures in the paper can be reproduced by the code stored in plot_paper.py. One can either use the prepared data (saved in the directory data) or produce those data anew.